MLX90621

16x4 IR array

Datasheet

39001090621

Page 40 of 44

Datasheet IR16x4

Rev 3.0

15 September 2016

FAQ

15.

When I measure aluminum and plastic parts settled at the same conditions I get significant errors on aluminum.

Why?

Different materials have different emissivity. A typical value for aluminum (roughly polished) is 0.18 and for plastics

values of 0.84…0.95 are typical. IR thermometers use the radiation flux between the sensitive element in the sensor

and the object of interest, given by the equation

Where:

Ë is the the Stefan-Boltzmann constant,

and the object temperature that we need.

Note that the temperatures are all in Kelvin, heat exchange knows only physics.

When a body with low emissivity (such as aluminum) is involved in this heat transfer, the portion of the radiation

incident to the sensor element that really comes from the object of interest decreases – and the reflected

environmental IR emissions take place. (This is all for bodies with zero transparency in the IR band.) The IR

thermometer is calibrated to stay within specified accuracy – but it has no way to separate the incoming IR radiation

into real object and reflected environmental part. Therefore, measuring objects with low emissivity is a very

sophisticated issue and infra-red measurements of such materials are a specialized field.

What can be done to solve that problem? Look at paintings – for example, oil paints are likely to have emissivity of

0.85…0.95 – but keep in mind that the stability of the paint emissivity has inevitable impact on measurements.

It is also a good point to keep in mind that not everything that looks black is “black” also for IR. For example, even

heavily oxidized aluminum has still emissivity as low as 0.30.

How high is enough? Not an easy question – but, in all cases the closer you need to get to the real object

temperature the higher the needed emissivity will be, of course.

With the real life emissivity values the environmental IR comes into play via the reflectivity of the object (the sum of

Emissivity, Reflectivity and Absorptivity gives 1.00 for any material). The larger the difference between

environmental and object temperature is at given reflectivity (with an opaque for IR material reflectivity equals 1.00

minus emissivity) the bigger errors it produces.

After I put the MLX90621 in the dashboard I start getting errors larger than specified in spite that the module was

working properly before that. Why?

Any object present in the FOV of the module provides IR signal. It is actually possible to introduce error in the

measurements if the module is attached to the dashboard with an opening that enters the FOV. In that case portion

of the dashboard opening will introduce IR signal in conjunction with constraining the effective FOV and thus

compromising specified accuracy. Relevant opening that takes in account the FOV is a must for accurate